Introduction

Most large and medium-sized organisations have shifted the focus of some or all of their organisational software development from tailored software, developed in object-oriented programming languages, to the use of configurable off-the-shelf packages or ERP systems. These systems offer a range of built-in functionality that can be adapted to specific requirements and offer opportunities for rapid system deployment and faster 'time-to-value', where the 'time-to-value' is the time after deployment when the system is fully operational and delivering value to the company.

However, there are real problems in introducing these systems as they require extensive process changes so that the organisation's processes are aligned with the assumptions in the software. Sometimes these are straightforward but, more commonly, process change is more difficult than anticipated, new systems are deployed later than planned and may not be used to their full capability.

Based on extensive experience of organisational studies of software systems and software engineering, Ian Sommerville will describe socio-technical systems engineering, an approach which can help managers and software engineers understand some of the organisational issues that can lead to software problems. Socio-technical systems engineering can help identify potential problems that might arise when new systems are introduced and can shorten the time required to deliver real value to the organisation. This approach has been designed for incremental adoption with little or no additional risk to the organisation. Although the illustrative examples will be drawn from COTS systems, STSE is useful in any situation where new software leads to business process changes.

Principal Themes

The tutorial has three principal themes:

1. The notion of time to value. Delivering and deploying software quickly is often seen as the principal objective of many software projects. In this talk, I will argue that time to value rather than time to deployment is what we should consider to be most important. This recognises that, immediately after deployment, most new systems do not deliver value to an organisation but usually lead to reduced efficiency. Over time, as the software becomes part of the business processes, it becomes more effective and, some time after deployment, delivers real value to the organisation. This time is the time to value.

2. Socio-technical systems. I introduce the notion of socio-technical systems and suggest that for organisational systems, we should be concerned with socio-technical systems engineering rather than simply software engineering. I discuss what is meant by socio-technical systems engineering, what is involved and why it can contribute to reducing time to value.

3. Getting started with socio-technical systems engineering. In this part of the tutorial, I explain how you can get started with socio- technical systems engineering and discuss how even a relatively small commitment to this process can be effective in reducing time to value.

Who should attend?

Managers and engineerings involved in planning, procuring, specifying and implementing new organisational software, either using COTS/ERP systems or through a conventional software development process.

-- back to Top --

Introduction

Evolving from traditional software components, Web services have received extensive studies within the software engineering community. Web services transcend traditional distributed object technologies (e.g. RPC, CORBA, DCOM, etc.) in that programming interfaces are described, published, discovered, executed, and composed using open standards such as XML, SOAP, WSDL, UDDI, BPEL, etc. Standardisation represents ‘a shared understanding towards implementation independence’, a key element for building loosely-coupled distributed applications through interoperability. However, with the recent development of Web services specifications, some scholars argue that “Web services do not have much in common with the Web”. It appears that Web services technology offers far more advantages in a non-Web environment.

Moreover, the current Web has undergone a radical change towards a highly user-centred platform where end users can easily participate and collaborate for their own benefits and profits. The great success of Web2.0 applications (e.g. Flickr, Youtube, Del.icio.us, etc.) indicates that the Web community has already adapted itself to accommodate massive user participation, which in turn provides astonishing reward. Nevertheless, Web services communities that enable services “on the Web”, appear reluctant (if not unaware) of these radical changes. Industry vendors and standards organisations continue pushing complex WS-* specifications. Researchers have applied much existing work – ontology engineering, artificial intelligence, workflow theory, agent technology, peer-to-peer infrastructure, etc., – to facilitate dynamic service discovery, execution, orchestration and coordination. However, to our best knowledge, little research has been formally carried out to justify or utilise the ‘Web’ part of ‘Web services’.

In this tutorial, we will review contemporary research that aims to ‘align’ the Web services with the modern Web in two significant aspects – Web services architectural styles and Web services discovery mechanism. Based on a comprehensive survey, we will identify several critical open issues that need to be addressed when incorporating Web (2.0) techniques and philosophies into the Web services practice. These key issues and their associated solutions give rise to our proposed research framework – An Alternative Web Services Paradigms. This tutorial will also introduce our recent work in this direction: (1) An evolution trend driven by engineering principles for Internet-scaled Web services architectural design. (2) A Web services discovery platform built on top of Web 2.0 techniques and attitude. In summary, this tutorial demonstrates how to move towards creating an alternative Web services paradigm that aligns itself with the modern Web.

Who should attend?

Researchers, executives, and software engineers, who are planning to leverage emerging Web technologies into Web services projects in order to create scalable and user-centred Web services applications.

-- back to Top --

Introduction

This tutorial gives an introduction to wireless sensor networks (WSNs) and how they can be utilised and integrated in an industrial setting. The basic principles presented will be supplemented with results from extensive laboratory tests and studies of such networks. The authors have for several years been working on evaluating wireless sensor systems, examining specifications and testing WSN technologies in laboratories. Experiences from this work will be presented, in addition to a real-world example from an offshore oil production plant in the North Sea. This includes software interfacing towards the existing monitoring system and network infrastructure.

Motivation

In an Oil & Gas production platform there can be tens of thousands of sensors that are used for measuring different physical quantities such as temperature, pressure, flow, vibration etc. Usually, only 10-20% of these sensors are connected to the monitoring and control systems. For the rest of the sensors, a field operator has to manually read the measurement value of each sensor and upload it into the backhaul systems. This labour is time-consuming and leads to delayed, irregular and extended intervals between monitoring, as well as erroneous readings. WSNs have the potential to simplify sensor deployment and installation, thus reducing the number of un-connected sensors. By completely eliminating the need for wires, battery operated wireless sensors offers a cost-efficient and easy way to deploy sensors in existing installations. However, there is still the challenge of integrating the new sensor technologies into existing monitoring and control systems.

Who should attend?

Anyone who wants a general introduction to wireless sensor networks, and who wants to learn how they can enable flexibility and cost reductions related to the use of sensor monitoring and control in industrial plants and facilities.

-- back to Top --